Texas Instruments expands its microcontroller product portfolio and software ecosystem to help edge AI land in various devices

Integrate TinyEngine ™  NPU's new MCU has joined Texas Instruments' comprehensive AI hardware, software, and tool portfolio to assist engineers in deploying intelligent technology to various applications.

TI is currently integrating TinyEngine NPU ™  Integrated into its entire microcontroller product portfolio, including general-purpose and high-performance real-time MCU.

News Highlights:

Compared to similar microcontrollers (MCUs) without accelerators, TI's integrated TinyEngine ™  The neural processing unit (NPU) can achieve a maximum delay reduction of 90 times and a single inference energy consumption reduction of over 120 times when running AI models.

TI's newly launched universal and real-time MCU integrates TinyEngine ™  NPU， Can bring more efficient edge AI capabilities to various system applications ranging from simple to complex.

With the help of TI CCStudio ™  The integrated generative AI in the IDE, along with over 60 models and application examples provided by CCStudio Edge AI Studio, allows developers to quickly and easily add edge AI functionality to any device.

Recently, Texas Instruments (TI) launched two new microcontroller (MCU) series with edge artificial intelligence (Edge AI) capabilities, fulfilling the company's commitment to implementing edge AI throughout its entire embedded processing product portfolio. MSPM0G5187 and AM13Ex MCU integrate TI's TinyEngine ™  Neural Processing Unit (NPU) is a hardware accelerator designed specifically for MCUs that optimizes deep learning inference operations, reducing latency and improving energy efficiency during edge processing.

TI's embedded processing product portfolio is supported by a comprehensive development ecosystem that includes the CCStudio integrated development environment (IDE). Its generative AI feature supports engineers to accelerate code development, system configuration, and debugging through industry standard agents and models, paired with TI data, using simple language. TI is promoting the popularization of edge AI in various electronic devices, with applications ranging from real-time monitoring in wearable health monitors and household circuit breakers, to physical AI functions in humanoid robots. These end-to-end innovative achievements will be exhibited at the TI booth at the 2026 International Embedded Exhibition held in Nuremberg, Germany from March 10th to 12th.  

TI Embedded Processing and DLP ®  Amichai Ron, Senior Vice President of Products, said, "About 50 years ago, TI invented digital signal processors (DSPs), laying the foundation for today's edge AI processing. Now, TI is leading the next stage of innovation by integrating TinyEngine ™  NPU is integrated into our entire microcontroller product portfolio, covering both general-purpose and high-performance real-time MCUs. By fully empowering AI in our software, tools, devices, and ecosystem, we are committed to making edge AI accessible, making it easy for every customer and application to use edge AI functionality. ”

Bob O'Donnell, President and Chief Analyst of TECHnalysis Research, a market research firm, pointed out that although many focuses are on AI acceleration and NPU for large SoCs, in fact, some of the more interesting and far-reaching applications in the AI field can be achieved on small chips such as microcontrollers. Edge side AI acceleration applications can make consumer electronic devices smarter and industrial devices more efficient. In addition, if these chips can be combined with software tools that also use AI to assist development, the powerful advantages of AI acceleration can be brought to a wider range of engineers and device designers

Advanced intelligence within reach

From fitness wearable devices to household appliances and power systems, consumers always expect daily technology products to be more intelligent. However, due to high costs, power consumption requirements, and programming barriers, many engineers believe that AI functionality is exclusive to high-end applications. TI's new MSPM0G5187 Arm ®  Cortex ®- The M0+MSPM0 MCU has brought significant transformation to embedded designers, enabling them to incorporate edge AI into a multitude of simpler, more compact, and cost-effective applications.

Through local computation, TinyEngine ™  NPU can work in parallel with the main CPU running application code to perform calculations required by neural networks. Compared to similar MCUs without accelerators, this hardware acceleration can:

Minimize flash memory usage to the greatest extent possible.

The maximum reduction in single AI inference latency is 90 times.

The energy consumption of a single AI inference is reduced by more than 120 times.  

Such high efficiency enables resource constrained devices, including portable battery powered products, to easily handle AI workloads. The MSPM0G5187 MCU provides an affordable alternative to other MCU or processor architectures, effectively reducing system and operational costs.

Real time control and AI acceleration for multi motor systems

In the application of motor control in household appliances, robots, and industrial systems, more and more systems require intelligent functions such as adaptive control and predictive maintenance, but in the past, implementing these functions required complex multi chip designs. With C2000 ™  The real-time MCU product line has been leading in the field of motor control for over 20 years. TI is the first in the industry to launch a high-performance Arm Cortex-M33 core TinyEngine ™  The new AM13Ex MCU integrates NPU and advanced real-time control architecture on a single chip.

This highly integrated design enables designers to simultaneously implement complex motor control and AI functions without the need for external components, thereby reducing bill of materials costs by up to 30%. Key performance improvements include:

• Supports up to four motors to maintain precise real-time control loops, while allowing TinyEngine to ™  NPU runs adaptive control algorithms to achieve load detection and energy efficiency optimization.  

• The integrated trigonometric function mathematical accelerator has a computing speed 10 times faster than the implementation of Coordinate Rotation Digital Computer (CORDIC), providing more accurate and responsive motor control performance.

Easily train, optimize, and deploy AI models

Both newly launched MCU series support development using TI's free CCStudio Edge AI Studio. This development environment can simplify the model selection, training, and deployment process in TI's entire embedded processing product portfolio. This edge AI toolchain provides engineers with high flexibility to run AI models on TI MCUs through hardware or software implementation. At present, the tool has provided over 60 models and application examples, which can help developers quickly deploy edge AI on any device, and more tasks and models will be added in the future.

Texas Instruments unveiled at the 2026 International Embedded System Exhibition

During the 2026 International Embedded Exhibition, TI showcased how its technology can help engineers accelerate development speed with AI, improve performance through edge AI, and deploy AI functionality at the edge of factories, buildings, and automobiles. In addition, the exhibition also highlighted TI's partner program and network, which provides a solid foundation for innovative embedded solutions to be brought to market faster.

Packaging and supply situation

• The production version of MSPM0G5187 MCU is now available on TI.com.

• AM13E23019 MCU is now available in pre production version. Other packaging and memory models will be released before the end of 2026.   

• Supports multiple payment and shipping methods.